The present invention relates generally to a device for warning of abnormal or undesirable conditions of an automotive engine or vehicle equipment in the form of human voice announcements which are stored in a recording medium. More specifically, the invention relates to an improvement in the voice warning device so that it permits the passengers to listen to warnings repeatedly when the passenger fails to understand when first reproduced.
In the prior art, there have been developed various device for warning of abnormal or undesirable conditions of the engine or vehicle equipment by vocal recordings in response to sensor signals which are generated when the sensor detects the abnormal or undesirable condition. Typically, states to be monitored are key removal when the driver gets out the vehicle, extinguishing headlights when the ignition key switch is turned off, fuel supply decreasing below a predetermined level, incomplete closing of the passenger doors, and so on. Therefore, the sensors for detecting the condition of key switch, lighting switch, fuel level, and door switch have been provided for the vehicle. Each sensor detects the condition of one state to be monitored and produces a command signal when it detects an unwelcome state to make the reproducing means operative. In order to obtain exact information, it may be preferable to repeat the warning when the passenger fails to understand while the warning is performed the first time.
Conventionally, such voice warning devices having repeat mechanisms have been known. Such devices include a digital processor serving as a voice synthesizer or a tape recorder for storing the voice warning for the preselected items and means for reproducing the warning announcements. However, both the synthesizer system and the tape recorder system are expensive to apply to the vehicle.